FP7 project - SAM.SSA

Transcription

1 Sugar Alcohol based Materials for Seasonal Storage Applications FP7 project - SAM.SSA Sugar Alcohol based Materials for Seasonal Storage Applications Workshop and Onsite Demonstration CiCenergigune Miñano, Alava, Spain

2 Session 1, Sub Session: Heat Transfer Enhancement Presented by: Alain CELZARD Contributions by: Prasanta JANA Vanessa FIERRO Alain CELZARD

3 Context Sugar alcohols: xylitol, sorbitol, mannitol, erytritol, etc., and their blends: molecular alloys Phase-change materials are investigated for thermal energy storage from the 70 s, but Challenge = seasonal storage

4 Context Phase-change materials are investigated for thermal energy storage from the 70 s, but Challenge = seasonal storage Sugar alcohols: xylitol, sorbitol, mannitol, erytritol, etc., and their blends: molecular alloys High energy density: compact storage systems High undercooling : long-term storage Possible crystallisation by thermal shock or other means: heat recovery at the desired moment Poor thermal conductors

5 Context Open porosity higher than 80% Thermal conductivity within the range 3 10 W/m/K Cost less than 6 /kg Wide pore size for easy and in-depth impregnation with SA Average thermal conductivities wanted! A very unusual request Why?

6 Context Need of a medium conductivity related to the optimisation between: - Short time for heat delivery - High power delivery Crystallite growth velocity Viscosity

7 Context Effect of the thermal conductivity of the porous host on delivered power:

8 What kind of porous matrices? Porous carbon matrices have many advantages : Cheaper than metal foams Are much lighter than metal foams Can have green precursors Can be prepared with any porosity Can have a broad range of physical properties Can be functionalised and/or nanostructured Require high-temperature treatment under inert atmosphere

9 What kind of porous matrices? Most common carbon foams Reticulated vitreous carbon (RVC) foams (Derived from thermoset polymers) Extremely low thermal conductivity Made from non-renewable resources Expensive Duocel RVC foam (ERG Corporation) Cellular graphitic carbon (CGC) foams (Derived from pitch or coal extracts) Very high thermal conductivity Made from non-renewable resources Prohibitive cost POCOfoam graphite foam (ORNL)

10 What kind of porous matrices? CVC and RVC foams Made from cheap (possibly renewable) resources Low cost Low thermal conductivity Cellular Reticulated Strategy : Preparing composite carbon foams containing a filler with a high thermal conductivity

11 Two approaches Autogenous foaming of sucrose solutions Sucrose + Graphite powder + Water + Nickel nitrate Stirring Sucrose-graphite resin 120 C, 48 h Solid green foam Carbon foam Carbonisation (900 C) Template method based on PU foams PU foam Polymeric foam cured at 150 C Carbon foam pyrolysed at 1000 C

12 Carbon foams from sucrose Thermally induced decomposition and resultant foaming Decomposition of a nitrate whereas a sucrose solution is polymerized in acidic conditions, leading to a foamed carbon precursor obtained at 120 C. Sucrose + Water + HNO 3 and/or metal nitrates + graphite Viscous resin, starting foaming 120 C in Teflon beaker, 48 h Solid foam 120 C, 48 h 900 C, 2 h Composite carbon foam

13 Carbon foams from sucrose Experimental design Reasons: - Need to maximise both porosity and thermal conductivity, whose changes are antogonistic - Save time, cost and materials Principles : - Choosing the most relevant parameters to be changed for preparing as few samples as possible - From the experimental results and the corresponding modelling, being able to predict the properties of any other formulation Implementation: - Choice of a graphite and a catalyst leading to the best results: a natural graphite having the lowest available particle size on one hand, and nickel nitrate on the other hand - Use of Design-Expert software for optimising the amounts of all ingredients, using porosity and thermal conductivity as responses.

14 Context Some results Thermal conductivity Porosity From the model: R1= A B C AB AC BC R2= A B C AB AC BC If A = 20, B = 6, C = 1: R1 = (model) versus 3.1 W/m/K (experiment) R2 = (model) versus 85.8% (experiment)

15 Carbon foams from sucrose Same procedure applied to much bigger samples : Slightly different R1= A B C AB AC BC R2= A B C AB AC BC but more favourable results : less graphite and less nickel are required for getting the same results! If A = 80, B = 20, C = 2: R1 = (model) versus 3.1 W/m/K (experiment) R2 = (model) versus 85.8% (experiment)

16 Carbon foams from sucrose One step further : more constraints for maximising both conductivity and porosity Sucrose Graphite Ni nitrate Conductivity Porosity Case # A B C R1 R2 1 Minimum Minimum Minimum Maximum Maximum 2 Maximum Maximum Maximum Maximum Maximum 3 In the range In the range In the range Maximum Maximum 4 In the range Minimum Minimum Maximum Maximum Examples : Cost is not important Graphite and Ni need to be minimised

17 Stress (MPa) Carbon foams from sucrose Mechanical properties of optimised foams : 10 9 (a) 8 7 (b) 6 5 MPa (big foam) MPa (small foam) Strain (%) Composite foam Compressive strength (MPa) Elastic modulus (MPa) Small sample Big sample

18 Carbon foams from sucrose Hydrophobisation of optimised foams : Different techniques used : - Direct chemical fluorination Foams of overall composition CF 0.04 CF Electrochemical grafting of fluorinated groups Si - Surface coating with silica by sol-gel method O O O Si Si O O O O HO - Chemical grafting of perfluoroalkyl chains

19 Carbon foams from sucrose Effect of hydrophobisation on subcooling : Some foams promote nucleation (no hydrophobic effect), others hinder nucleation (hydrophobic)

20 Carbon foams from sucrose Conclusion : Composite, graphitised, carbon foams made from sucrose and natural graphite are : - cheap ( 3 /kg) - highly porous (86%) - easy to impregnate with sugar alcohols - moderately conducting before ( 3 W/m/K) and after ( 3.5 W/m/K) impregnation with sugar alcohols - resistant (3 MPa) and stiff (100 MPa) before impregnation - even more after impregnation: 10 et 200 MPa, respectively - promote or delay heterogeneous nucleation, depending on surface functionalisation

21 MASA Thermal Conductivity Improvement WP 4 (MASA Thermal Conductivity Improvement) Presented by: Dr. Mani Karthik Work Package Leader: Dr. Alain Celzard Participants: Dr. Bruno D Aguanno and Dr. Abdessamad Faik

22 Graphite Composite Foam by Template Method Tasks: Synthesis of graphite composite foam Structural characterization Impregnation of the matrix with PCMs (Erythritol) Thermophysical characterization Chemical stability of the PCMs with carbon matrix Target Porosity: % Density: g/cm 3 Thermal conductivity: 2-4 W.m -1.K -1 Cost: 6-7 /Kg Mechanical stability Hydrophobic Results Achieved Porosity: % Density: g/cm 3 Thermal conductivity: 2-4 W.m -1.K -1 Cost: 3-7 /Kg Mechanical stability : MPa Hydrophobic: contact angle 120 o

23 Preparation of Graphite Composite Foams

24 Graphite Composite Foam Surface Morphology - SEM Properties of the Carbon Foams Target: Porosity: %

25 Compressive stress-strain behavior of the carbon foams Samples Carbon-Graphite foam (CGF1) Carbon-Graphite foam (CGF2) Graphite- Graphite foam (GGF1) Composition Polymer/Graphite/Ni (wt. %) Compressive strength (MPa) Elastic modulus (MPa) 60/40/ /50/ /40/ Reticulated Vitreous Carbon (RVC) Commercial Ultramet carbon foam ( MPa) Commercial ERG carbon foam ( MPa) Compression Strength ( MPa)

26 Surface Functionalization of Carbon Foams - Hydrophobic Carbon Foam Water droplet on the surface of the carbon foam Contact Angle Measurement EDAX Mapping Contact Angle = 120 o Graphite Foam Surface coating with polytetrafluoroethylene (PTFE) Surface coating with tetraethyl orthosilicate (TEOS)

27 Preparation of Erythritol-Graphite Composite Foam Sugar Alcohol Infiltration The erythritol infiltration was performed by placing a piece of carbon foam at the bottom of a pool of liquefied erythritol in a beaker/glass plate. After infiltration, the foam piece was removed from the beaker and allowed to cool down. The excess of erythritol in the outer surface of the carbon foam was removed by carefully scraping with a blade or knife. Impregnation of Sugar Alcohol Infiltration of Sugar alcohol is ca wt % depends on the porosity of the carbon foam

28 Structure of Graphite Foam Before and After Erythritol Impregnation Before impregnation After impregnation

29 Characteristics of the Materials Mechanical Stability Thermal Stability

30 Contact Angle Measurement Graphite composite foam Hydrophobic Erythritol-Graphite Composite Foam Hydrophilic

31 Chemical Stability of the PCM within the Carbon Matrix Compatibility between Erythritol and Graphite Foam XRD Patterns In-situ XRD Patterns (In-situ heating chamber)

32 Thermal Conductivity Enhancement of PCM Composite Laser Flash Apparatus (LFA 457, Netzsch) k =..C p where k is the thermal conductivity in W/mK is the thermal diffusivity in mm 2 /s is the density in g/cm 3 Cp is the specific heat in J/g.K

33 Small Experimental Device

34 Conclusions Carbon-graphite and graphite-graphite composite carbon foams were prepared by template method using commercially available polymeric foams (PU) as macroporous sacrificial template. Erythritol-graphite foam as a stable composite PCM was obtained by a simple incipient wetness impregnation method with high impregnation ratio. The in-situ XRD result clearly indicates that there is no chemical reaction between erythritol and graphite and thus the composite has not only good structural stability but also good compatibility between erythritol and graphite composite foam. It was found that the thermal conductivity of the erythritol-graphite composite foam (3.88 W/mK) was enhanced ca. 5 times higher as compared with that of pristine erythritol (i.e. 0.7 W/mK). This enhancement can significantly reduce the charging and discharging times of the PCM storage system. The PCM/foam composite has a melting point of 118 o C and latent heat of ca. 250 J/g which corresponds to the mass percentage (ca. 75 wt. %) of the erythritol within the composite foam. It can be concluded that the carbon matrix is not affecting the melting behaviour and enthalpy of the erythritol.